As we move through the biotechnology age fostered by the human genome project a premium has been placed on the development of high throughput methodologies to obtain and analyze sequence information. To meet this demand, the multifunctional DNA microarray platform has gained notoriety leading to an explosive growth in application methods using the same.
More importantly, the evolution of world events and the emergence of bioterrorism in mainstream society have led to a growing sentiment amongst the scientific community and lay people alike that new, rapid, and accurate techniques for biological threat identification and eradication must be developed. The concept of a microarray used for broad-spectrum pathogen identification has considerable and obvious appeal to both medical practice and national defense. It is within this framework that the present inventors have endeavored.
Heretofore, for the purpose of pathogen identification, approaches generally rely on the ability of immobilized “probe” DNA sequences on the surfaces of microarrays to hybridize with complementary genomic “target” that is uniquely identifying of a particular category or specific strain of microbial pathogen. Various microarray technologies have been developed for this purpose, varying in the density of probes and the time ranges required for assay completion.
One technical challenge for pathogen detection with microarrays arises due to the difficulty in obtaining samples with a sufficient quantity of pathogen nucleic acid. Thus, for a majority of sample types, some sort of target amplification will likely be required to provide sufficient copies of pathogen gene markers for detection by microarray hybridization. Unfortunately, conventional methods for this amplification do not scale well in comparison to the number of probes that can be placed on a microarray chip. However, the most commonly employed means of providing sufficient quantities of genomic target to detect hybridization relies upon genotypic identification methods that utilize molecular biology-based techniques, such as the polymerase chain reaction (PCR). These techniques offer several potential advantages over conventional microbiological approaches. Nucleic acid amplification strategies base pathogen identification on the detection of genetic information contained within the organism, such that culturing the organism is not required.
Although PCR-based assays are sensitive, accurate, and rapid, these methods also introduce a new set of problems. As successful identification depends almost entirely on appropriately chosen primer sets, as PCR-based testing requires assumptions about the exact sequences pertaining to the identity of the target organism(s). Consequently, there is a critical need for advanced diagnostic systems that can detect both assumed and unanticipated pathogen sequences. DNA microarrays, which enable the simultaneous interrogation of thousands of genetic elements, address this crucial need. Here, the term “microarray” refers to any type of planar substrate or solid beads presenting a high multiplicity (102 to 106) of individual sites, each presenting nucleic acid probes designed to selectively capture complementary strands of target (i.e. pathogen or host) nucleic acid.
However, the majority of pathogen identification microarrays described in the literature is prepared using oligonucleotides that are robotically spotted onto derivatized glass surfaces (typically 3×1 inch microscope slides). This approach allows the most flexibility with regards to the size of the oligonucleotides that are deposited, ranging from 20-mers to cDNA PCR products of several thousand base pairs (bp). With few exceptions, the detection event is an increased level of fluorescence originating from a spot following hybridization of a fluorophore-labeled target nucleic acid.
Short (14-25 mer) oligonucleotides, immobilized inside acrylamide pads, have been applied extensively to pathogen identification (Strizhkov et al., 2000; Vasiliskov et al., 1999) in a collaborative effort between Argonne National Lab (DOE, USA) and the Engelhard Institute of Molecular Biology (Moscow, RU) under the leadership of Andrei Mirzabekov. In addition, low-density microarrays (several hundred features per 3×1 inch microscope slide) have been used for determination of drug resistance determinants (Volokhov et al., 2003). One distinguishing aspect of this body of work is the use of three-dimensional polymer matrices for probe immobilization instead of two-dimensional planar surfaces.
More recently, Cherkasova et al have described the use of glass-immobilized short oligonucleotide spotted microarrays to map poliovirus mutations using overlapping 14-25 mer probes (Cherkasova et al., 2003). Two variations of this approach have been used: (1) Microarrays for Resequencing and Sequence Heterogeneity (MARSH) assay, and (2) Microarray Analysis of Viral Recombination (MAVR) assay. MARSH uses a set of overlapping (at half length) nucleotide probes for individual gene sequences. Hybridizations patterns allow the detection of single point mutations or substitution/deletion events to a resolution of half probe lengths (e.g. 7-10 bp) but does not allow for exact determination of position(s) or the nature of the mutation. Accordingly, conventional DNA sequencing technologies must be employed subsequently to determine these changes. MAVR uses organism-specific oligonucleotide probes that cover the entire genome at ˜150 nt spacings and is used to detect large scale genetic recombinations.
The DeRisi group at UCSF pioneered the use of long (70-mer) oligonucleotide probe microarrays for broad-spectrum pathogen identification (Wang et al., 2002; Wang et al., 2003). The use of long (70 nt) oligonucleotides bears implicit advantages and disadvantages. One advantage is that higher degrees of sensitivity can usually be achieved with 70-mer probes compared to shorter ones (e.g. 20-25 mers). However, specificity is reduced because 70-mer target/probe hybridizations are generally insensitive to significant numbers (e.g., 7-10) of single base mismatches, whereas shorter probes provide much greater sequence specificity.
DeRisi's group described the use of spotted microarrays having 1,600 different 70-mer oligonucleotide probes to identify a variety of viruses responsible for common respiratory infections (Wang et al., 2002). The probes were selected for each pathogen using an algorithm that located discriminatory sequences from a list of known viral genomes. A serial combination of a previously described (Bohlander et al., 1992) method and subsequent PCR/Klenow fragment-based amplification was used to achieve non-biased amplification of both viral RNA and DNA, allowing generation of sufficient amounts of target amplicons for successful microarray hybridization and detection via fluorescent label. (N.B. This protocol was placed into the public domain via the DeRisi lab website (http://derisilab.ucsf.edu)). The time required from sample preparation to obtained result was approximately 24 hours. Because exact sequence information was not attainable from such arrays, pathogen identifications were made on the basis of a hybridization pattern that could be empirically determined for each pathogen or strain. In a related report from the same group (Wang et al., 2003) similar microarrays were prepared using highly conserved sequences in an effort to capture as many microbial species as possible from a sample. Following physical removal of the pathogen sequences from the microarray, the sequences are cloned and sequenced using conventional DNA sequencing technologies. No measure of analytical/clinical sensitivity or specificity for pathogen detection in clinical specimens was provided in the work from the DeRisi group.
In contrast to the above-mentioned approaches using spotted microarrays, Affymetrix, Inc. (Santa Clara, Calif.) uses high-density probe fabrication technology to construct “tiled” microarrays using 4 probes each in both the sense and anti-sense directions for each nucleotide base to be resequenced. Thus, single base substitutions are directly detected by the hybridization pattern (for additional information see Affymetrix CustomSeq design manual). Several groups described the use of tiled microarrays for pathogen genotyping. (Kozal et al., 1996) utilized this type of microarray to measure mutational drift in HIV while Gingeras et al (Gingeras et al., 1998; Troesch et al., 1999) used a tiled array of 65,000 oligomer probes to resequence and accurately identify 70 clinical isolates of 27 mycobacterial species and 15 rifampin-resistant M. tuberculosis strains. More recently, Andersen et al. (Wilson et al., 2002b) described the use of tiled Affymetrix microarrays for the identification of biological warfare agents. Their approach relied entirely on the use of specific PCR reactions performed in parallel to generate sufficient pathogen target DNA for microarray hybridization. In all cases listed above, specific PCR primers were used to amplify DNA targets prior to microarray hybridizations, through the use of conserved primer sites, or in the work of Andersen et al. (Wilson et al., 2002a; Wilson et al., 2002b), by carrying out up to ˜150 different PCR reactions in multi-well format and pooling the amplicons.
U.S. Pat. No. 6,228,575 B1 describes the same data as described by Gingeras (Gingeras et al., 1998) and Troesch (Troesch et al., 1999). In this patent, which is incorporated herein by reference in its entirety, target pathogen sequences are tiled onto arrays. Because several types of variations (esp. insertion/deletion or frequent multiple substitutions) in pathogen sequence can perturb hybridization patterns, Gingeras et al. used differential measures of specific pathogen hybridization patterns to identify individual mycobacterial variants. That is, identifications required a priori knowledge of a differential hybridization pattern that was empirically determined in ground truth experiments.
As stated above, there is a critical need for advanced diagnostic systems that can detect known and pathogen genomic sequences as well as variations of those sequences. More particularly, there remains a critical demand for DNA microarray techniques that are fast and reliable, but are free from the systemic bias flowing from the specific PCR based methods that have heretofore been employed.